• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Watercool Heatkiller V Pro RTX 4090 Strix/TUF GPU Block

VSG

VSG

Editor, Reviews & News
Staff member
Joined
Jul 1, 2014
Messages
3,725 (0.94/day)
We begin a new GPU water block roundup with the recently released Heatkiller V Pro RTX 4090 Strix/TUF block from Watercool. It uses a new cooling engine tweaked for ultimate performance and is part of an excellent design which is functional while looking gorgeous too.

Show full review
 
Watercool makes great products ! A warm recommendation
 
When installing the block, I had to chuckle a little when I saw that even the power connector gets a thermal pad:
PXL_20230806_131949046.jpg


For the black version, I recommend checking out the stainless steel side cover for a two-tone look.
PXL_20230806_140639752.png
 
I've always thought Watercool's Heatkiller line is the best looking on the market
 
If the heat is being dissipated elsewhere (radiator), idk why the block has to be 25% longer than it needs to
 
The plate is exactly as long as needed only covering the important parts delimited by the MOSFETs on both sides. The acrylic however covers the whole PCB and extends the channels to the edge. In that sense it could be modified to look more like a CPU block and indeed 20% smaller if the acrylic is cut to the exact shape of the plate but this would be purely cosmetic doesn't take away the cooling elsewhere. 435 grams of copper is like 4$ so yeah.
 
Do you know roughly what the average power draw was when you were doing the time spy extreme test?

It's obviously a VERY different setup, but, I have the 4090 FE block from watercool. I used Kingpin KPX paste (not the included Kryonaut), had roughly a 0.5gpm flow rate and my ambient was between 22-23c. Card was OC'd to around 3K/11.6K on the core/vram, max power limit etc. When I did a 30 minute loop of Port Royal 4K the highest delta peak I saw was 18c at ~515w. After 15 minutes I measured the average power draw at around 470w and the average delta was around 16c. This was all read through HW info.

Half the flow rate should nullify any advantage from the TIM. Did you measure the core temp Delta via a physical thermistor or through software ?How much of a difference is there between physical temperature monitoring and software for GPU? Even with all the different variables, it would seem like hw info would be off by quite a lot.
 
Thanks for this review, very apreciated, and very nice product. I've the EK block fort the 4090FE but it's possible that i'm gonna take another 4090 and this could be the right block.

Anyway can i make a stupid question? Are we sure that measuring the delta from liquid temp it's the right way to judge theese blocks? Because the liquid temp could change even because of a better thermal transfer, so maybe the delta from ambient could be better, no?
 
I am a big fan of heatkiller water blocks. I personally used a HK block designed for 775 socket for a long time. I even added new wholes to make it compatible with a newer platform, and used it like that till I sold out mobo cpu ram kit to someone else.
Heatkiller blocks remind me of Koolance which I always chose when they were still available.
My only regret is that if you check at their compatibility lists, their blocks are designed for the cards selling in Europe but not always USA.
This was my big disappointment when I tried to use their older G{U block with gtx 1xxx. I could not find a card it would fit.
But their design is something I give highest rating. Additional compliments for threats not being made in very fragile clear acrylic material. I fear such design greatly after losing multiple parts to cracks.
I hope to get back to Heatkiller when rtx 5000 comes out. Possibly a combo of CPu and GPu waterblocks. Aquacomputer radiator would also be nice completing this german setup.
 
You must log in or register to reply here.
Back
Top